Crosscorrelation Migration of Free-Surface Multiples in CDP Data

1
Crosscorrelation Migration of Free-Surface
Multiples in CDP Data
Jianming Sheng University of Utah February, 2001
2
Outline

• Objective
• Crosscorrelation migration
• Numerical examples
• Summary

3
Objective
To image free-surface multiples by
crosscorrelation migration
To improve the migration image quality by
attenuating the artifacts caused by free-surface
multiples.
4
Outline

• Objective
• Crosscorrelation migration
• Numerical examples
• Summary

5
Principle of CCM
6
Principle of CCM
Imaging Condition
7
Asymptotic Analysis
8
Asymptotic Analysis
Under stationary phase condition
Correct image

Artifacts
Negligible
9
Asymptotic Analysis
Crosscorrelation migration can migrate the
multiples to the correct position but generate
artifacts as well CCM image alone can not give
the reflectivity distribution!
10
Key Idea of CCM
CCM Image
Kirchhoff Image
11
Key Idea of CCM
12
Outline

• Objective
• Crosscorrelation migration
• Numerical examples
• Summary

13
Numerical Examples

• Three-layered model
• Nine-layered model
• SEG/EAGE salt model

14
Three-Layered Model
Model
CCM Image
100 200 300 400 500
100 200 300 400 500
15
Three-Layered Model
Kirchhoff Image
Product Image
100 200 300 400 500
100 200 300 400 500
16
Nine-Layered Model
Model
CCM image
500 1000 1500 2000 2500
500 1000 1500 2000 2500
17
Nine-Layered Model
Kirchhoff Image
Product Image
500 1000 1500 2000 2500
500 1000 1500 2000 2500
18
SEG/EAGE Salt Model
0
600
1200
1800
Depth (m)
2400
3000
3600
19
CCM Image
0
600
1200
1800
Depth (m)
2400
3000
3600
20
Kirchhoff Image
0
600
1200
1800
Depth (m)
2400
3000
3600
21
Product Image
0
600
1200
1800
Depth (m)
2400
3000
3600
22
All right, multiples are migrated, well, is
it useful?
Yes.
23
Kirchhoff Image
0
600
1200
1800
Depth (m)
2400
3000
3600
24
Product Image
0
600
1200
1800
Depth (m)
2400
3000
3600
25
Outline

• Objective
• Crosscorrelation migration
• Numerical examples
• Summary

26
Summary

• Multiples can be considered as signal and
  correctly imaged by the crosscorrelation
  migration

• By multiplying the crosscorrelation and Kirchhoff
  migration images, the true reflectors can be
  enhanced and the artifacts can be attenuated.

27
Further Work

• To attenuate the artifacts generated by CCM

• To deal with the high-order multiples and
  internal multiples.

28
Acknowledgment
I thank the sponsors of the 2000 University of
Utah Tomography and Modeling /Migration (UTAM)
Consortium for their financial support .